Problem with Nested Dielectrics?

[pixelpeepH7GXH]

It appears to me that there is a problem with the new Nested Dielectrics. I have made a quick test scene with a Dark Green Glass shape, with something akin to Red Wine inside (Geometry of the wine extended to overlap the glass inside polygons). As I have always done, I give the Glass a shallow Transmission Depth and the Wine a relatively larger Transmission Depth. I will attach 3 screen grab renders of the Glass by itself, the Wine by itself and another with the Wine in the Glass. As you can see the Wine basically disappears leaving only some specular details behind. I have a link to the C4D file as well if you want to check out. I would be curious to hear what you think of it and if I am doing something wrong or if it is a bug. BTW. I am using C4D R21. I have not tested it on other versions. Also this is using CPU ... if I use the GPU it looks like the render where Nested Dielectrics is turned off.

C4D R21.207, C4DtoA-3.2.1

Scene saved to Dropbox below.

https://www.dropbox.com/s/7d08asr1ba9cjx1/Dielectric_Nesting_TestScene.c4d?dl=0

Log file below

https://www.dropbox.com/s/77jfhi9yna3ly7p/LogFile.zip?dl=0

[pixelpeepH7GXH]

It appears to me that there is a problem with the new Nested Dielectrics. I have made a quick test scene with a Dark Green Glass shape, with something akin to Red Wine inside (Geometry of the wine extended to overlap the glass inside polygons). As I have always done, I give the Glass a shallow Transmission Depth and the Wine a relatively larger Transmission Depth. I will attach 3 screen grab renders of the Glass by itself, the Wine by itself and another with the Wine in the Glass. As you can see the Wine basically disappears leaving only some specular details behind. I have a link to the C4D file as well if you want to check out. I would be curious to hear what you think of it and if I am doing something wrong or if it is a bug. BTW. I am using C4D R21. I have not tested it on other versions. Also this is using CPU ... if I use the GPU it looks like the render where Nested Dielectrics is turned off.

C4D R21.207, C4DtoA-3.2.1

Scene saved to Dropbox below.

https://www.dropbox.com/s/7d08asr1ba9cjx1/Dielectric_Nesting_TestScene.c4d?dl=0

Log file below

https://www.dropbox.com/s/77jfhi9yna3ly7p/LogFile.zip?dl=0

[Anonymous]

I'm using mtoa and I think it's broken too. The overlapped geometry doesn't seem to show up

[Anonymous]

I'm using mtoa and I think it's broken too. The overlapped geometry doesn't seem to show up

[Anonymous]

should work if you put the sphere inside the liquid with :
glass:3
liquid:2
sphere:1

[Anonymous]

should work if you put the sphere inside the liquid with :
glass:3
liquid:2
sphere:1

[pixelpeepH7GXH]

I agree... "it should work"... but it doesn't. And it should not require a 3rd item to make it work.

Anyone know how to submit a bug report other than here?

[pixelpeepH7GXH]

I agree... "it should work"... but it doesn't. And it should not require a 3rd item to make it work.

Anyone know how to submit a bug report other than here?

[thiago.ize]

Thanks for the scene. This is indeed a bug in Arnold which we are in the midst of fixing.

[thiago.ize]

Thanks for the scene. This is indeed a bug in Arnold which we are in the midst of fixing.

[pixelpeepH7GXH]

Thank you Thiago.

[pixelpeepH7GXH]

Thank you Thiago.

[pixelpeepH7GXH]

Still broken with C4DtoA-3.2.0.1-win-R21

[pixelpeepH7GXH]

Still broken with C4DtoA-3.2.0.1-win-R21

[thiago.ize]

c4dtoa 3.2.0.1 is still using the same version of arnold core 6.1.0.0 (see release notes https://docs.arnoldrenderer.com/display/A5AFCUG/3.2.0.1) so it won't contain any new core fixes. If the arnold core gets updated and you still see problems, please let us know.

[thiago.ize]

c4dtoa 3.2.0.1 is still using the same version of arnold core 6.1.0.0 (see release notes https://docs.arnoldrenderer.com/display/A5AFCUG/3.2.0.1) so it won't contain any new core fixes. If the arnold core gets updated and you still see problems, please let us know.

[Anonymous]

I can confirm the same bug in Houdini 18.5.351 / HtoA 5.5.0

I'm glad I found this post.. I was going nuts 🙂

[Anonymous]

I can confirm the same bug in Houdini 18.5.351 / HtoA 5.5.0

I'm glad I found this post.. I was going nuts 🙂

[pixelpeepH7GXH · Accepted solution]

Thank you 'Oh great makers of Arnold'!.

It seems to be fixed in the Arnold 6.1.0.1 core used in C4DtoA 3.2.1.

Cheers

[pixelpeepH7GXH · Accepted solution]

Thank you 'Oh great makers of Arnold'!.

It seems to be fixed in the Arnold 6.1.0.1 core used in C4DtoA 3.2.1.

Cheers

[theo7R7XM]

Hi,

I'm going crazy here too...

I'm using maya 2022 & Arnold Core 7.1.3.0 (but same issue with two previous version)

When I set a higher dielectric priority for the glass than for the red liquide: the liquid disappear. (image 2.jpg)

When I set a lower dielectric priority for the glass  than for the liquide: it's render fine. (image 1.jpg)
Should it not be the opposite ?

[theo7R7XM]

Hi,

I'm going crazy here too...

I'm using maya 2022 & Arnold Core 7.1.3.0 (but same issue with two previous version)

When I set a higher dielectric priority for the glass than for the red liquide: the liquid disappear. (image 2.jpg)

When I set a lower dielectric priority for the glass  than for the liquide: it's render fine. (image 1.jpg)
Should it not be the opposite ?

[theo7R7XM]

oh my bad, i found the solution.
My understanding is that Nested dielectrique need to have transmission_depth > 0 to work.
Per the documentation:

"Ensure that transmission_depth is set to a non-zero"

[theo7R7XM]

oh my bad, i found the solution.
My understanding is that Nested dielectrique need to have transmission_depth > 0 to work.
Per the documentation:

"Ensure that transmission_depth is set to a non-zero"